Sie sind auf Seite 1von 5


, 1946



1. Ultraviolet absorption spectra measurements are recorded for aqueous solutions of Dglucosyl-N-butylamine and are compared with those for si:milar solutions of D-glucose in alkali. 2 . The above data are interpreted as demons trating that the incipient coloration appearing

when aqueous solutions of D-glucosyl-N-butylamine are allowed to stand a t room temperature, is caused by hydrolysis of the compound followed by the action of the liberated hydroxyl ions upon D-glucose. It is probable that this is a general reaction of sugar-amino condensation products in these relatively high pH ranges.







The Willgerodt Reaction. 11. A Study of Reaction Conditions with Acetophenone and Other KetoneslTa
BY DELOSF. D E T A RAND MARVIN ~ CARMACK Although the Willgerodt reaction4 has been applied to a large number of ketones for the preparation of carbonamides, no systematic study of the effect of variations of reaction conditions appears t o have been published. The attempt of Willgerodt and Scholtz5 to replace the aninioniuni polysulfide reagent with colorless ammonium sulfide gave poor results. One useful variant of the original procedure, introduced by Fieser and Kilmer,6 involves the addition of dioxane to the reaction mixture; the presence of the organic solvent generally permits the use of lower reaction temperatures, and yields are often greatly improved. I n an effort to improve the usefulness of the Willgerodt synthesis of amides and t o throw light on the reaction mechanism, we have made a study of the effect of the time, temperature, composition of the reagent, and presence of organic solvents on the reaction of acetophenone to form phenylacetamide. As a result of this study a modified procedure was developed which gave a total of 85.67&of phenylacetamide and phenylacetic acid as compared with the total of 63% by the original procedure of Willgerodt and Merk. The improved methods were applied also to the reactions of a number of other ketones according to the general equation


The results are summarized in Tables I1 and I11 in the Experimental.

(1) For l.he previous paper of the series, see Cavalieri, Pattison and Carmack, THIS JOURNAL, 1783 (1945). 67, (2) From the Ph 11. Dissertation of DeLos F. DeTar, University of Pennsylvania, 1944. (3) Present address: Chemistry Department, Cornel1 University, Ithaca, N. Y . (4) Willgerodt, Ber., 20, 2467 (1887); 21, 534 (1888). (5) Willgerodt and Scholtz, J . prakf. Chcm., [2] 81,382(1010). (6) Fieser and KJmer, TEISJOURNAL, 1354 (1940); a hibliog62, raphy of earlier references to the Willgerodt reaction appears in this paper. (7) Willgerodt and Merk, J . prakt. Chcm., 121 80, 102 (1909).

According to the classical procedure the ammonium polysulfide reagent is prepared by dissolving a n excess of sulfur in concentrated ammonium hydroxide which has been previously saturated with hydrogen sulfide. Although the published procedures are rather indefinite the molecular proportions of reactants have probably been of the following order: ketone (1))ammonia (7), hydrogen sulfide (4)and sulfur (2). Using a standardized procedure for small-scale runs with acetophenone, we determined that the temperature range of 160-190 gives the highest yields within a convenient reaction time of four to six hours. Using combinations of sulfur, ketone, and analyzed solutions of ammonium sulfide-ammonium hydroxide in varying proportions, we found that the highest yields of amide were obtained when a large excess of sulfur was used; the optimum ratio appeared to be approximately 10 gram-atomic weights of sulfur for each mole of ketone. When the concentration of ammonium hydroxide and the proportions of sulfur and ketone were held approximately constant and the concentration of hydrogen sulfide added to the starting mixture was varied, the interesting result emerged that the best yields of amide were obtained when only low concentrations of hydrogen sulfide were added initially. Higher concentrations of hydrogen sulfide in the reagent markedly lowered the yields of amide. This finding suggested the desirability of omitting hydrogen sulfide entirely from the reagent, a modification which gave very satisfactory results with acetophenone. Runs in which only concentrated ammonium hydroxide, sulfur and acetophenone were present initially produced higher yields than runs using comparable proportions of the same reagents with added hydrogen sulfide. It is apparent from a consideration of the stoichiometry of the Willgerodt reaction that the over-all process involves a net oxidation which is almost certainly a t the expense of the added



Vol. 68

sulfur. Soirie hydrogen sulfide is therefore formed disk consisting of a coarse fritted glass disk was found to increase of solution. as a reaction product, and will be prescnt during tliffereiit the rate concentrationsStock solutions containing relative of ammonia and hydrogen most of the reaction period whether m y is added sulfide were made up and analyzed in order to give a range to the mixttre initially or not. I t is iiot surpris-- i f concentratioris for the experiments with acetophcnonc. ing therefore that the results obt:iiiietl with low The concentration of ammonia in the various solutions was (0-1 -11) initial concentratioris of Iiydrogen sill- determined by titration with standard hydrochloric acid, methyl ~ - d indicator. as of sulfide fide were n:)t greatly tliffereiit frotii those ob- tisiiig deterniiiictl rby a procedure Concentrations described wre similar to that tained with concentrated aniiiioiiiuiii hytlrositl(~ Iiy Treadwell and Hall.3 The colored solutioiis formed by niixitig ammonium and no added h y d r q e n sulfide. I t is not known whether hydrogen siilficle pluys hydroxide, hydrogen sulfide, and sulfur are extremely complex mixtures contailling many molecular and ionic an essential role in the course of the IVillgerotlt species and a number of equilibrium systems which are reaction. If i t is :I iiecessary coinpoiient of the iiot well understood. In order to express the composition reaction mixture the cridence iiitlicntcs clearly of our reageiits in a simple, if somewhat empirical, manner, that the amount of hytlrogen sulfide generated by ire have recorded the total molarity of ammonia and molarity of solutions of ammothe reaction itsel1 is sufiicient in the case of aceto- nium sulfide, hydrogeu sulfide in the the titrations, withas determined directly in phenone. This does n o t appear to be true for all out attempting to evaluate the relative amounts of these types of ketones sirice it. was shown that the best substances with respect t o their salt forms. Immediately before use in a reaction, a known volume yields of amide were obtained from purely aliof analyzed ammonium sulfide solution was mixed phatic ketones when the reagent initially con- a weighed quantity of finely powdered sulfur to formwith the tained enough hydrogen sulfide to convert all of polysulfide reagent. Since hydrogen sulfide was not always used and since consequently the sulfur was not the ammonium hydroxide to atiimoniuiii sulfide. With alkyl aryl ketones derived from phenaii- always completely in solution, it seemed preferable not sulfur in terms of molarity threne, pyrenc, biphenyl and similar polyc~clic to express the amount of added polysulfide. Instead, the or an empirical formula for hydrocarbons, a reagent containing approxi- ratio of gram-atoms of sulfur to moles of ketone was remately 0.7 111 Iiydrogen sulfide (or the equivalent corded. This value has been referred to as the mole in atnmonitiiii sdfidej gave very sntisfactory ratio of sulfur in the subsequent discussion. Procedure results and is convenient to prepare. The pres- -Redistilled for the Small-Scale Runs with Acetophenone. acetophenone (Eastman Kodak Co.) (1.000 ence of low concentrations of sulfide greatly- in- g.) was weighed into a 5-inch soft-glass test-tube, the apcreases the sdubility of sulfur in conipnrisoii with propriate weight ( t 0 . 0 2 9.) of flowers of sulfur was added the solubility in aiiinioniurn hydroxide, ant1 i t is and 5.00 ml. of analyzed ammonium sulfide of the desired composition used, possible that: the increase in the solubility of sul- was pipettedwas introduced. Organic solvent, when tubes into the mixture at this point. The fur partially- offsets the tendency of hydrogen were sealed and heated in a gas-heated Carius furnace, the temperature of which was held within + 5 . Usually sulfide t o decrease the yields of amide. A study of the effect of the concentration of a number of tubes in a given series were heated simultaneously in the furnace so that results would be ammonia 011 the yield of phenylacetamide from parable in spite of possible fluctuations in the directly comtemperature. acetophenone showed that the yields increased Unless otherwise specified, the time of heating was four with increasing concentration and leveled off in hours, not counting the time required to reach the stated the range of 12 to 15 moles of ammonia (or ani- temperature or cool to room temperature. The tubes were allowed to cool t o room temperature in moniuni salt) per liter of reagent. the furnace. In experiments with acetophenone no residThe effect of eth:iiiol, pyridine or dioxane added ual pressures were observed after the tubes had cooled. to the rextioii inistures was iiivestigntecl. The contcnts of each tube were evaporated to dryness in Ethanol iniprovctl thc [vicltls slightly, but pyri - n 400-ml. beaker on a water-bath to remove volatile anilnonium sulfide, leaving as a residue phenylacetamide, dine and tliosaiic~ wet-(: iiiorc cffecti\-r. Ihcrc sulfur, ammoniutn phenylacetate, and small amounts of is nct evidence that orgaiiic solvelits take Itart i i i 1)y-products. It was determined that losses of amide by the cheiriical reactions, :im1 i t is :issunlet1 that volatilization were negligible. The residue in each beaker their effect is rluc t o tlie iiicrcasetl solubility of the was extracted with six 10-m1. portions of boiling water, the hot solutions charcoal, starting ketones in the iiiisture of organic solvent :mtl evaporated to were clarified withflask. The filtered, dryness in a tared crystalj and aqueous reqcri t.. I i i runs w th cert:ki acyl line residue was found to consist of phenylacetamide with derivatives o the liiglicr polycyclic hydrocarbons very small amounts of phenylacetic acid and ammonium a practical atlvant;tge in the u s e of dioxane rather phenylacetate. The values in duplicate runs showed good than pyridinc was uotctl. The amides often reproducibility. of Optimum Time and Temperature.Determination sep:tr:itc>tl i i i fairly Inire crystalline stat? tlirectly I II a series of preliminary runs with acetophenone, follow~ 1 frioiii i l l ( - cc~ 1 1 ~n.;icLioii iiiistures coiitxiiiing itig the procedure described above, it was determined that

diusclnc, \ ) t i t

t t ~ i ~ i x i ~ iiiostlv i i t i t d




pyridine rnixtures, iircessitatilig tiiethod oi isolation. Experimental


n i t

Reagents.--l3ydrogeii sulfide gas was passed with cooling into concentrated arrimoniunl hydroxide (15 M ) form to solutions of crdorless ammonium sulfide 5 A dispersing ___--.(8) As hydrogen sulfide di.sulves t h e t i t e r of amniutlid Id115 sloniy from approximately 1 .I! t o :ibout 12 I f :snil tlir ctmr?ntr:iticin ; sulfide rises grar1ii;illy t u n h u i 7 JI.

s t temperatures of 160 arid higher the nmximurn yield of amide was obtained in a heating period of t w o to threr hours, and that longer heating in most cases did not change tlie yield of amide greatly. A standard reaction time of four hours was adopted to ensure complete reaction without uunecessarily prolonging the period of heating and p-ornotirig hydrolysis of the amide. It is not certain that the runs at temperatures of 130 were complete within the four-hour periods, but at 160 and higher there seems little

Treadwell and Ifall, Analytical Clieiuistry, (<,I. I I , Johu TViley aud Sans, X e w \ork, N. Y., 1935. p. 630

Oct., 1946



reason to believe that longer heating would have increased the amount of amide. Effect of the Mole Ratio of Sulfur:Acetophenone.-The experiments shown graphically in Fig. 1 were made with an ammonium sulfide solution which was 12.8 AI in ammonia and 2.4 .I1 i n liydrogeri sulfide and to which varying proportioris of sulfur were added to give the mole ratios shown in the ab:,cissa The three curves in Fig. 1 represent, in ascending order, the results at 130 * 5', 160 * 5' and 190 f 5', respectively, all runs involving a heating period of four hours. The highest yield was obtained at 190" with a su1fur:ketone ratio of 10.

there is little change in the yield of amide, but at higher concentrations of sulfide the yields fall off sharply. Effect of the Initial Concentration of Ammonia.--The experiments shown in Fig. 3 were carried out a t 160" for four hours in concentrated arnmoniuin hydrouidc (15 M ) without hydrogen sulfide. T h e three curves represent, in ascending order, mole ratios of sulfur: ketone of 2, 5, and 10, respectively. The yields are seen to reach a maximum in the concentration range of reagent ammonium hydroxide (12-15 M). Under the conditions of these experiments a solution initially 1.67 A 1 in ammonia would contain exactly the theoretical amount of ammonia to convert all of the acetophenone to amide.

4 8 12 16 20 Mole ratio of sulfur: ketone. Fig. 1.--Effect of mole ratio of sulfur to ketone: 0, 130 * 5", 4, 160 =t 5'; 0, 190 * 5". All experiments with 1.000 g. of CeH6COCH3 and 5.00 ml. of ammonium sulfide solution (12.8 21 in ISH8 and 2.4 M , in H&) with 1 added sulfur as indicated; time, four hours.

0 4 8 12 1 6 Initial M of ammonium hydroxide. Fig. 3.-Effect of initial concentration of ammonium hydroxide: 0, S:CaHbCOCH3 mole ratio 2 : l ; A, S: CH ! C J 6 &O H mole ratio 5:l; 0 , S: CsHbCOCHs mole ratio 1O:l. All experiments with 1.000 g. of C6HjCOCH3,5.00 ml. of ammonium hydroxide of indicated M ,and sulfur in ratios shown; 160" for four hours.
Effect of Added Organic Solvents.-Table I shows the results of a series of experiments with 1-g. quantities of acetophenone and 5-ml. quantities of ammonium sulfide. In all runs except the one with dioxane the ammonium sulfide contained 12.8 M ammonia and 2.4 ill hydrogen sulfide, with su1fur:ketone ratios as shown; 2 . 5 nil. of

Effect of Initial Concentration of Hydrogen Sulfide.The experiments shown in Fig. 2 were made a t 160' for four hours. The initial concentration of ammonia i n the ammonium sulfide solution was in every case within the range 11.3-11.9 M , the concentrations of hydrogen but sulfide varied froin 0 to slightly over 5 . The three curves, in ascending order, represent sulfur :ketone mole ratios of 3, 5, and 10, respectively. It can be seen that for initial concentrations of hydrogen sulfide in the range 0-1 M

Mole ratio sulfur: ketone

Organic solvent added

Yield of phenylacetamide, %


Initial JI of hydrogen sulfide. Fig 2.---Effect of initial concentration of hydrogen sulfide: 0 , S:CsHaCOCHs mole ratio 2 : l ; A, S:C&I5COCH? mole ratio 5:l; E, S:C6HbCOCH3 mole ratio 1 0 : l . All experiments with 1.000 g. of C6HsCOCH3,5.00 ml. of ammonium sulfide solution (11.3-11.9 Jf in SH3 and as indicated in H P S ) ,and sulfur in t h r ratios shown; 160" for four hours.

130 * 5" 130 * 5" 130 * 5" 130 * 5" 130 * 5' 130 f 5" 158 * 5' 158 * 5' 158 * 5" 158 =t 5' 158 * j3 158 . j 0 158 * < 5 O

5 3 3 10 12 12 5 3 5 3 10
12 12

Sone Ethan ol Pyridine


14 13 27 4 *5
i ,1 9IJ

Ethaiiol Pyridiiir Sone Ethanol Dioxane Pyridine None Ethanol Pyridine


60 i6







Vol. 68

ethanol or pyridine was added, where indicated. In the experiment with dioxane, the ammonium suliide was initially 12.3 M in ammonia and 0.69 M in hydrogen sulfide, and 2.0 ml. of dioxane was used. The best yield was obtained with a sulfur ratio of 12, a t 158 * 5",in the presence of added pyridine. Phenylacetamide from Acetophenone (Preparative Procedure).-Acetophenone (25.0 g., 0.208 mole), 37.5 g. of sulfur (1.17 gram atomic weights), 50 ml. (0.76 mole) of concentrated (15 M ) ammonium hydroxide, and 30 ml. of pyridine were sealed in a 22-mm. 0. d. hard glass (not Pyrex) bomb tube and heated to 157 * 6' for approximately four and one-half hours. The tube was allowed to cool to room temperature and opened (no residual pressure). The reaction mixture was transferred to an evaporating dish and evaporated to dryness on a water-bath. The dry residue of sulfur and reaction product was leached with a total of about 500 ml. of boiling water in several portions. The clarified filtrate was cooled, whereupon 20.0 g. of phenylacetamide, m. p. 156-158" cor., separated from the solution. Concentration of the filtrate yielded a second crop of 2.7 g. of amide. The filtrate from the second crop was evaporated to dryness and extracted with ether, yielding an additional 0.32 g. of amide and 1.2 g.

a mixture of 2,4- and 2,5-diphenylthiophenes prepared by heating phenylacetylene with sulfur.lO Reactions of Alkyl Phenyl Ketones.-Several straightchain alkyl phenyl ketones were treated according to the modified Willgerodt procedure using concentrated ammonium hydroxide, sulfur, and pyridine under conditions found t o be optimum for acetophenone, i. e., approximately 5 ml. of concentrated ammonium hydroxide, 2.66 g. of sulfur and 2.5 ml. of pyridine for each gram of ketone, at 165' for four to five hours. The products were isolated as described for phenylacetamide, and were purified by recrystallization from appropriate solvents. In one case the amide was converted to the solid acid. The results are shown in Table 11. Reactions of Acyl Derivatives of Polycyclic Hydrocarbons.-Several acyl derivatives of polycyclic hydrocarbons were treated with a reagent prepared in the following manner. Powdered sulfur was suspended in ten times its weight of concentrated ammonium hydroxide and hydrogen sulfide was passed into the suspension with agitation just long enough to cause all of the sulfur to dissolve. Analysis of a typical solution showed that it contained 12.3 M of ammonia and 0.69 M hydrogen sulfide. In each of the experiments from one t o three grams of ketone was heated with the reagent and dioxane in a sealed glass tube at 160' for approximately six hours. The proporTABLE I1 tions were usually 5 ml. of polysulfide reagent and 4-6 ALKYL PHENYL KETONES THE WILLGERODT IN REACTION ml. of dioxane for each gram of ketone. The amide generally crystallized partially or completely upon coolM. P.. M. P., oc. % oc. ing, but additional product could be recovered after Amide (COT.) Yield (cor.) evaporation of the filtrate to dryness to destroy ammoKetone obtained amide amide acid nium sulfide. Some of the amides were hydrolyzed to the Propiophenone 6-Phenylpropionarnide 97 .,5- 98. 82 5a crystalline acids. The results of several typical experin-Butyrophenone y-Phenylbutyramide 83 - 84b 42 49-5OC ments are summarized in Table 1 1 Melting points are 1. n-Valerophenone 6-Phenylvaleramide 107. 5-10Sd 29 given for the purified specimens. It is t o be noted that the products in all cases had the a A sample prepared by another method melted a t 99.5100.2' cor. Reported m. p. 84.5.7 A specimen of the structures expected from the usual formulation of the acid, m. p 49-50' cor., was prepared by hydrolysis and Willgerodt reaction, i. e., no rearrangement of the carbon decarboxylation of ethyl 8-phenethylmalonate, then con- skeleton or loss of carbon atoms was noted, with one verted t o the amide, m. p. 83-83.5" cor., not depressed interesting exception: in the case of 9-propionylanthracene by admixture with the product of the Willgerodt reaction. the side chain was lost and anthracene was recovered in 85% yield. No similar example of the complete loss of c Reported m. p. 51", Fischer and Schmitz, Ber., 39, 2212 (1906). ** Reported m. p. 104-105", Eijkman, Chem. a side chain appears to have been noted previously in the Weekhlad, 5,655 (1908); Chem. Zentr., 79,II, 1100 (1908). Willgerodt reaction.


Retorie Amide obtained

M. p., "C. (cor.)


% Yield

M. p..




2-Acetylphenanthrene' %Phenanthreneacetamide* 247 -248 82 2-Propionylphenanthrene' P-(2-Phenanthrene)-propionamided 189.3-189.5" 66 176 -178' 4-Acetylbipheny1O 4-Biphenylacetamide' 242.5-243.5' 84 160.8-161.8' 9-Propionylanthracenek None' a Prepared by the procedure of Mosettig and van de Kamp, THIS JOURNAL, 3444 (1933). b A n d . Calcd. for 55, ClsHIJNO: N, 5.95. Found: N, 5.89, 5.87. Prepared by the procedure of Bachmann and Struve, THIS JOURNAL, 58, 1659 (1936). Anal. Calcd. for CI~HNNO:N, 5.62. Found: N, 5.65, 5.57. M. p. in an evacuated capill?. f Reported m. p. 177.2-178.4', Riegel, Gold, and Kubico, THIS JOURNAL, 1775 (1943); reported m. p. 177-177.5 , 65, Anal. Bachmann, ibid., 57, 1381 (1935). 0 Prepared by the procedure of Long and Henze, ibid., 63, 1939 (1940). Calcd. for C1,HlaNO: N, 6.63. Found: N, 6.51,6.56. M. p. in an evacuated capillary. i Reported m. p. 161-162", Lesser, German Patent, 658,114 (1938) ; C. A . , 33,4798 (1938). k Prepared by the procedure of Liittringhaus and Kacer, U. S. Patent 1,766,433 (1930). Cleavage to anthracene occurred (85% recovery). (4.2%) of phenylacetic acid. The combined yields of phenylacetamide and phenylacetic acid amounted to 85.8%. The acid, after crystallization from water, melted a t 76.3--77.3 cor. The residue of sulfur remaining from the original reaction mixture after extraction with hot water was extracted further with alcohol. A small amount of oil was isolated from the alcohol, and treatment of the oil with successive portions of ether, petroleum ether, and methan01 eventually afforded a few milligrams of powdery crystalline mat.eria1, m. p. 141-143 ' cor. Recrystallization of the solid from methanol gave pale yellow plates, m. p. 145-146 , which were insufficient for further purification. The material behaved, however, very much like

Summary A critical study was made of the effect Of time, temperature, composition of the reagent, and the presence of organic solvents on the yield of phenylacetamide from acetophenone in the Willgerodt reaction- A modified procedure the use of concentrated ammonium hydroxide, sulfur and pyridine without hydrogen sulfide was developed which gave higher y.ields than the classical reagent, ammonium polysdfide(10) Carmack and DeTar, THIS JOURNAL, 68, 2029 (1948).

Oct., 1946



The modified procedures were applied to the reactions of several straight-chain alkyl phenyl ketones and several acyl derivatives of polycyclic hydrocarbons. 9-Propionylan thracene was cleaved to give a

high yield of anthracene under conditions of the Willgerodt reaction.







The Willgerodt and Kindler Reactions. 111. Amides from Acetylenes and Olefins; Studies Relating to the Reaction I M e c h a n i s r n ~ ' ~ ~ ~ ~
BY MARVIN CARMACK DELOSF. DETAR' AND I n the course of our investigations of the Krillgerodt5 and Kindler6 reactions we have found that acetylenes and olefins7 behave similarly to ketones, forming carbonamides and thioamides in the two procedures, respectively. Thus, in analogy with tlie conversion of acetophenone to phenylacetarnide,2 phenylacetylene and styrene both give good yields of phenylacetamide optimum yields of amide from alkyl aryl ketones, the three types of compounds were found to produce amides in the following order of decreasing yields : ketones, acetylenes, olefins. Phenylacetylene and styrene react with morpholine and sulfur to produce phenylacetothiomorpholide. The yield of purified product is about the same in each case as that obtained from acetophenone by the procedure of Schwenk and Bloch6
C6Hb-COCHz CeHj-c~cHa C6Hs--CH=CHz

Paralleling the reaction of propiophenone, both I-phenylpropync and 1-phenylpropene yield pphenylpropionainide


+ S+
'N (C H Z C H P ') 0


'The unsaturated hydrocarbons thus share with the ketones the most remarkable feature of the Willgerodt. reaction-the disappearance of an unsaturated group and the oxidation of a terminal methyl group. Under standard reaction conditions which had previously been found2 to give
(1) EDITOR'S FoOTsOTE.-The original manuscript of this paper was received September 1.5, 1945. A manuscript by King and McMillan containing some closely related material and an interpretaf tion o the mechanism of the Willgerodt reaction was received September 22, 1945. I t was piauned that both papers would appear in JOURXAL. As the result of a clerical error in the same issue of THE) the Editor's Office, however, the paper by King and McMillan was JOUXNAL, 68, 633 (1946)) before the condensation published (THIS and revision of the p.%perby Carmack and DeTar had been completed, In view of tlie prior publication of King and McMillan's paper, the article by Carmack and DeTar has been further revised to eliminate some historical material and to call attention t o the results independently arrived a t by King and AfcMillan. (2) For the previous paper of this series see DeTar and Carmack,
THIS O U R X A L , 68. 2025 (1946). J

(3) From the Ph.D. dissertation of Debs F. DeTar, accepted by the Graduate Schuol of the University of Pennsylvania, 1944. (4) Present addresr.: Department of Chemistry, Cornell University, Ithaca, 3. Y . ( 5 ) For a bibliography of references to the Willgerodt reaction consult Fieser and Kiiiner, THIS JOURNAL, 62, 1364 (19401, and ref. 2. (6) For refprences t o the Kindler reaction, consult Schwenk and Blocb, ihid., 64, 3051 (1942). and ref. 15. (7) King and McMillan have observed independently t h a t olefins yield carbonamides under Conditions of the Willgerodt reaction; cf. footnote 1 and also ibid., 68, 525 (1946).

A marked evolution of hydrogen sulfides acconipanies the reaction of styrene. These reactions appear to be applicable to a wide variety of substituted acetylenes and olefins. We are exploring their scope and will report our findings in later publications. Several further experiments carried out in this Laboratory are of interest in connection with the development of a general theory of the course of the Willgerodt reaction. Phenylacetaldehyde gives phenylacetamide under the same conditions as acetophenone, phenylacetylene and styrene. P-Phenylpropionaldehyde and phenylacetone8 each give P-phenylpropionamide under the same conditions as propiophenone, 1-phenylpropyne, and 1-phenylpropene. However, methylphenylcarbinol,* P-phenylethyl alcohol,* /3-phenylethyl acetate, and methylbenzylcarbiriol fail t o yield appreciable amounts of amide when heated with a typical Willgerodt reagent a t 160". Using the ammonium polysulfide-dioxane reagent recommended by Fieser and Kilmer,5 we were able to confirm the formation of a-methyly-phenylbutyramide from isobutyl phenyl ketone, as reported originally by JVillgerodt and Merkg and recently reinvestigated by Fieser and Kilmer.
(8) Also investigated by King and McRIillan (footnotes 1 and 7 . ) (9) Willgerodt and Merk, J . puakl. Chcm , [ Z ] 80, 192 (19091.